Some thoughts ..
(a) aerodynamicists like measuring CG in %MAC for engineering reasons as handling qualities are linked conveniently to %MAC for design and flight test work. Apart from TP/FTE types, who are interested in the same aspects, the piloting fraternity doesn't need to worry too much about the subtleties.
(b) the Flight Manual and loading data generally are prepared by the OEM's aerodynamics section
(c) guess what .. the aerodynamicists take the easy way out and prepare the AFM/loading manual data using %MAC rather than refiguring it in pilot friendly CG arm values.
(d) for the operator, %MAC is just a different, more complicated, way of saying the CG is so many units aft of whatever datum you choose.
The two are related by a simple equation
%MAC = ((CG measured from datum - LEMAC referenced to datum)/length MAC)*100
The original post identified the reasons for having the %MAC overlaid on the trimsheet. Depending on the aircraft, an actual CG near either the forward or aft limit and the stab set for the opposite end of the envelope can result in some interesting times for the pilot on the day ... so it augurs well to get it right at the sums stage ..
Is one better than the other ? ... with my occasional aerodynamicist/flight test hat on, %MAC is better .. but, with a routine pilot hat on .. give me inches or mm aft of the datum every time, please ..